Authors: Zhang Song-lin; Li Xue
Addresses: School of Electronic Communication Engineering, Henan Institute of Technology, Xinxiang, 453002, China ' School of Electronic Communication Engineering, Henan Institute of Technology, Xinxiang, 453002, China
Abstract: Echo state networks (ESN) are an emerging learning technique proposed for generalised single-hidden layer feed forward networks (SLFNs). However, the conventional ESN ignores training data timeliness, which may reduce prediction accuracy for time varying data. To solve this problem, a novel algorithm based on ESN with adaptive forgetting factor (AF-ESN) is proposed. The adaptive forgetting factor is introduced to ESN sequential learning phase, which automatically tunes the valid training data window size according to prediction error magnitude. A comparison of the proposed AF-ESN with other algorithms is evaluated on three chaotic time series and an actual time series. Compared with conventional ESN and FOS-ELM (online sequential extreme learning machine with forgetting mechanism), though AF-ESN consumes much computation time, AF-ESN provides the highest prediction accuracy with high stability.
Keywords: ESN; echo state networks; chaotic time series; neural networks; SLFNs; single-hidden layer feed forward networks; short wave radio transmitters; Moore-Penrose; training data timeliness; time varying; forgetting factor; regularisation factor; time series prediction; adaptive forgetting; sequential learning.
International Journal of Intelligent Systems Technologies and Applications, 2017 Vol.16 No.1, pp.80 - 93
Available online: 29 Dec 2016 *Full-text access for editors Access for subscribers Purchase this article Comment on this article